Flashcards in 1 - Intro Deck (27)
What are the possible hidden reasons for medical visits? (PAILS)
What is the time-honored principle for patient care?
Diagnose first and then treatment
What circumstances would be appropriate to treat prior to making a diagnosis?
Emergencies or Subacute conditions (like GERD, where improvement in symptoms based on H2 blocker would be diagnostic)
What are the 3 principal clinical problem solving methods?
2) Algorithmic (includes Heuristic)
True of False: Diagnostic testing is often more reliable than clinical opinion.
The number of cases of a particular dz within a defined population at a given point in time.
The rate of occurrence of new cases of a disease or condition during a given time interval
baseline probability of disease before the results of a diagnostic test is known
Change in disease probability based on a positive or negative test.
So what will you tell your patient about their likelihood of having a disease based on both the pre-test probability and their actual test results?
True or False: Tests with higher sensitivities and specificities will also have high positive and negative predictive values.
How do you calculate sensitivity?
True positive results divided by the total number of patients who actually have the disease.
TP/(TP + FN)
How do you calculate specificity?
True negative results divided by the total number of patients who are actually disease free.
How do you calculate positive predictive value?
True positive results divided by the total number of patients who tested positive (regardless of actual disease state)
How do you calculate negative predictive value?
True negative results divided by the total number of patients who tests negative (regardless of actual disease state)
The sensitivity and specificity of a test are combined into a single entity called ______________
the likelihood ratio (LR)
How do you calculate the LR for a TRUE positive? (How likely is it that a positive test result is a true positive rather than a false positive?)
The positive likelihood ratio is the sensitivity divided by one minus the specificity
LR+ = sensitivity/(1 - specificity)
Calculate the LR+ for a test whose sensitivity is 0.92 and whose specificity is 0.82.
Is this a good test?
Good tests will have results >>1, so yes, this is a good test.
How do you calculate the LR for a FALSE negative?
(How likely is it that a negative test result is wrong and the patient actually has the disease?)
The negative likelihood ratio is one minus the sensitivity divided by the specificity.
LR- = (1 - sensitivity)/specificity
Calculate the LR- for a test whose sensitivity is 0.92 and whose specificity is 0.82.
Is this a good test?
Good LR- tests will have results <<1, so yes, this is a good test.
Why are likelihood ratios better than predictive values?
Because the results of predictive values will vary based on the prevalence of disease
What are the 3 main reasons for ordering laboratory tests?
2) identify or confirm dx
3) monitor dz status
If you suspect a disease and you'd like to diagnose it definitively, use a test with high ____________. However, if you'd like to rule out a diagnosis, use a test with high______________.
What 4 major factors influence a clinician's decision to treat?
3) confidence in dx
4) seriousness of disease
What is the term for the "threshold" at which a clinician will chose to actively treat a patient?
(Threshold meaning: are you 70% confident, 85% confident, etc.... that your patient has a given disease?)
"benchmark for action"
Which treatment decision making model assumes that the physician is best qualified to make decisions for the patient and does not place very much, if any, responsibility on the patient?
Which treatment decision making model assumes that the informed patient is best qualified to make decisions for themselves? Here the physician answers all questions and provides ample information but does not express clinical preferences and ultimately leaves decisions to the patient.